A camera-IMU system extrinsic parameter calibration method

摘要

V-SLAM technology (simultaneous localization and mapping using visual and inertial sensors) has achieved more and more attentions from industry and academic fields. It can provide the 3-D motion of a vehicle when no external references (e.g., GPS) are available. The camera-IMU extrinsic calibration parameters can significantly affect the estimation precision when the V-slam algorithm is performed. Because the translation of the camera-IMU sensor can be measured easily manually, the paper is mainly concerned about its attitude parameter. We propose a camera-IMU calibration method which can get the attitude between the camera and IMU. We capture some chessboard images on the ground in different attitudes and different positions of the sensor. The algorithm can achieve the camera-IMU extrinsic calibration parameters with less computation complexity and less time consuming. The experimental results validate the proposed method and qualify its accuracy.